Imagine having a virtual copy of your body that updates in real-time as you live your life. This digital double helps doctors predict how a stroke patient will react to specific diets or medicines without guessing. It acts like a high-tech GPS for healthcare, constantly recalculating the best treatment path as new health data comes in.
Current stroke care is fragmented and relies on intermittent data updates, leading to outdated treatment decisions. Doctors lack tools to simulate how a specific patient will respond to changes in diet or medication over time.
A platform featuring a Personal Data Vault and hybrid digital twins that combine mechanistic models with machine learning to provide continuous patient stratification.
If you are an eHealth developer dealing with static patient data that quickly becomes outdated — this project developed a platform that connects various apps to a living digital twin. This allows your software to provide updated diagnosis and treatment assessments every time new patient data is produced.
If you are a pharmaceutical company dealing with unpredictable patient responses to stroke medications — this project developed hybrid mechanistic and ML models. These simulate patient-specific responses to drugs at an intracellular and whole-body level, ranging from seconds to years.
If you are a hospital network dealing with fragmented data across prevention, acute care, and rehab — this project developed a Personal Data Vault system. This integrates the entire patient journey into one backend, supporting personalized medicine for 300 prevention study participants and a rehab pilot.
Based on available project data, no specific pricing or commercial cost model is mentioned; the project is funded by an EU contribution of EUR 5,698,475.
The project describes a scalable platform designed to connect multiple apps and 8 partner hospitals, suggesting a design intended for wide deployment across healthcare systems.
Based on available project data, specific licensing terms are not listed, but the system uses a Personal Data Vault where data is controlled by the patient.
It uses semantic harmonization and federated learning to securely re-train models using data from cohort databases without compromising security.
The project period runs from 2023-05-01 to 2027-04-30, indicating that full validation and testing will continue until April 2027.
The consortium is well-balanced for a translation project, consisting of 15 partners across 9 countries. With a 20% industry ratio (3 companies, including 3 SMEs), there is a clear path toward commercialization, while the 10 academic and research partners provide the deep technical expertise needed for the hybrid AI and mechanistic modeling.
Contact Linköpings Universitet regarding the STRATIF-AI platform architecture.
Contact us to explore licensing opportunities for the digital twin backend.
